Home › Blog › Tests and assessment
The CAT4 Test Explained: What It Measures, What the Scores Mean and Why Year 7s Sit It
The CAT4 test, short for Cognitive Abilities Test, Fourth Edition, is the reasoning test many secondary schools give new pupils in the first weeks of Year 7, and some primary schools use earlier. If your child has just sat it, or you have been sent a report full of numbers, this page explains it in plain terms. A score of 100 is exactly average for a child of that age; it is not a percentage. CAT4 measures how a child reasons in four different ways, not what they have been taught, and schools use it to spot strengths, hidden difficulties and pupils who may be doing less well than their ability suggests. It needs no preparation, and it is not an IQ test.
What the CAT4 test is
CAT4 is published by GL Assessment. It is taken on screen or on paper, at nine levels of difficulty matched to age, from Level X for Year 2 pupils through Pre-A and A to G, with Level D the usual paper for Year 7. GL says it is standardised on 25,000 pupils and checked every year against results from around a quarter of a million.
From Level A upwards the test comes in three parts of about 45 minutes each, usually spread over a day or two. The two youngest levels have two parts of about 30 minutes. Everything is multiple choice.
The four batteries
CAT4 is made up of four reasoning areas, which GL calls batteries:
- Verbal reasoning: thinking with words, such as spotting how words relate. It matters most for subjects heavy in reading and writing.
- Quantitative reasoning: using numbers to solve problems, without relying on taught maths methods.
- Non-verbal reasoning: solving problems with shapes and patterns, important in maths and science.
- Spatial reasoning: picturing and manipulating shapes in three dimensions, the skill behind a lot of STEM work and one that school tests rarely measure.
Because the questions test reasoning rather than knowledge, CAT4 can pick up ability that a child's classwork or SATs results do not show, for example a pupil who reasons well but struggles to read, or one who is still learning English.
What the scores mean
Each battery produces three scores, plus an overall mean.
Standard Age Score (SAS)
The main number. It adjusts your child's raw score for their exact age and compares it with a national sample of children of the same age. The average is 100. Roughly two-thirds of children score between about 85 and 115. A score of 100 is not 100%; a score of 112 is not 112%. The mean SAS is simply the average of the four battery scores.
National Percentile Rank (NPR)
The share of children of the same age who scored the same or lower. An NPR of 84 means your child did as well as or better than 84% of children their age. An NPR of 50 is average.
Stanine
A simpler 1 to 9 scale. Stanines 4, 5 and 6 are average and between them cover 54% of children; 7 to 9 are above average and 1 to 3 below. GL recommends stanines when talking to parents because they are easy to read and do not look like IQ scores.
| Stanine | Description | Standard Age Score | Share of pupils |
|---|---|---|---|
| 9 | Very high | 127 and above | 4% |
| 8 | Above average | 119 to 126 | 7% |
| 7 | Above average | 112 to 118 | 12% |
| 6 | Average | 104 to 111 | 17% |
| 5 | Average | 97 to 103 | 20% |
| 4 | Average | 89 to 96 | 17% |
| 3 | Below average | 82 to 88 | 12% |
| 2 | Below average | 74 to 81 | 7% |
| 1 | Very low | 73 and below | 4% |
Learning profiles: verbal bias and spatial bias
The report also compares your child's verbal and spatial scores and places them in one of seven profiles, from an extreme verbal bias through no bias to an extreme spatial bias. Neither direction is better. A child with a verbal bias tends to learn well through reading, discussion and writing. A child with a spatial bias often learns best from diagrams, models and visual explanations, and may find essay-based subjects harder than their overall ability would suggest. Teachers use the profile to adjust how they teach, not to sort children into groups by ability.
How schools use CAT4
- Getting to know a new intake. Year 7s arrive from many primary schools with different records. CAT4 gives every pupil the same baseline. GL also suggests a second test in Year 9 or 10 for an updated picture before GCSE.
- Setting. Some schools use CAT4 alongside SATs results and teacher judgement when placing pupils in sets. A good school treats it as one piece of evidence, not the only one.
- Spotting underachievement. When a pupil's CAT4 scores are well above their classwork or test results, it can point to a problem worth investigating, from a reading difficulty to unhappiness at school.
- Special educational needs. A big gap between batteries can be an early flag for a specific learning difficulty, though CAT4 does not diagnose anything. Our SEND support guide explains what to ask for if you are worried.
- Indicator grades. The reports include statistical indicators of likely GCSE and A-level grades. These are probabilities based on how pupils with similar scores have done nationally, not predictions or targets fixed to your child.
Some independent and international schools also use CAT4 in admissions. State schools in England cannot select pupils by ability, apart from grammar schools and a small number of partially selective schools. Some comprehensives do use a reasoning test for fair banding, which spreads places across the ability range rather than picking the highest scorers. For grammar school tests, see our 11-plus explained page for how those work.
Should my child prepare for CAT4?
No. GL Assessment says no preparation is needed, and schools generally ask families not to tutor for it. Practice can push a score up by making the question styles familiar, but that defeats the point: an inflated score can land a child in a set that moves too fast, or hide a difficulty the school would otherwise have spotted. A good night's sleep and breakfast beforehand do more good. If your child is anxious, it helps to tell them it is not a pass or fail test and nothing is decided on it alone.
Questions to ask the school
Schools do not have to share CAT4 results, but many send home GL's Individual Report for Parents or will talk you through the scores. Useful questions, especially at a Year 7 parents' evening:
- How have you used my child's CAT4 results, and alongside what other evidence?
- Does the gap between their scores suggest anything you are keeping an eye on?
- Are their current results in line with what CAT4 suggests they can do?
- Will they sit it again, and when?
For more on the move up to secondary, see our page on primary to secondary transition, or start from the Schools Insight homepage to compare schools near you.
Sources
GL Assessment, CAT4 product overview; GL Education support, CAT4 quick data guide and CAT4 quick start guide (levels and test timings).
Frequently Asked Questions
What is a good CAT4 score?
The average Standard Age Score is 100, and about two-thirds of children score between roughly 85 and 115. A score of 112 or more puts a child in stanine 7 or above, the top 23% for their age; 127 or more is the top 4%.
Is CAT4 an IQ test?
No. It measures reasoning in four areas against children of the same age, and the scores are not IQ scores. GL Assessment recommends schools report stanines to parents partly to avoid that impression.
Can you fail the CAT4 test?
No. There is no pass mark. It is used to understand how a child learns and to plan teaching, not to pass or fail anyone.
Should I tutor my child for CAT4?
GL Assessment says no preparation is needed, and most schools ask families not to practise. An inflated score can lead to the wrong set or hide a learning difficulty the test would otherwise have flagged.
When do children take the CAT4 test?
Most often at the start of Year 7, with some schools retesting in Year 9 or Year 10. Versions exist for children from about age 6, so some primary schools use it too.
What do verbal bias and spatial bias mean on a CAT4 report?
They compare a child's verbal and spatial reasoning scores. A verbal bias means the child is stronger with words; a spatial bias means they are stronger at picturing shapes and space. Neither is better; it shows how they may learn best.